{ "bench_version": 1, "track": "dynamics", "model": "rnn_small_phase_gru", "metric_direction": "lower is better", "n_seeds": 8, "baseline": { "best_cfg": { "lr": 0.006, "weight_decay": 0.0 }, "sweep": [ { "cfg": { "lr": 0.001, "weight_decay": 0.0 }, "mean": 0.002156580056180246, "std": 0.0005337133383512335, "per_seed": [ 0.0014283749042078853, 0.0019986501429229975, 0.0017361249774694443, 0.0022385285701602697, 0.0016417494043707848, 0.002327029127627611, 0.0030278568156063557, 0.002854326507076621 ] }, { "cfg": { "lr": 0.003, "weight_decay": 0.0 }, "mean": 0.0011022633989341557, "std": 0.00039374711633484005, "per_seed": [ 0.0009436382679268718, 0.0008971625356934965, 0.0006927313515916467, 0.001675509149208665, 0.0007120227674022317, 0.0008096328820101917, 0.0013870575930923223, 0.00170035264454782 ] }, { "cfg": { "lr": 0.006, "weight_decay": 0.0 }, "mean": 0.0007801908141118474, "std": 0.000154764005468824, "per_seed": [ 0.0007548112771473825, 0.0008833256433717906, 0.0005147535703144968, 0.0009292386239394546, 0.0005692947888746858, 0.0008575883111916482, 0.0007604749407619238, 0.0009720393572933972 ] } ], "full": { "mean": 0.0007801908141118474, "std": 0.000154764005468824, "per_seed": [ 0.0007548112771473825, 0.0008833256433717906, 0.0005147535703144968, 0.0009292386239394546, 0.0005692947888746858, 0.0008575883111916482, 0.0007604749407619238, 0.0009720393572933972 ], "n": 8, "signatures": [ {}, {}, {}, {}, {}, {}, {}, {} ] } }, "idea": { "best_cfg": { "lr": 0.006, "weight_decay": 0.0, "k": 1.0, "delta": 0.12 }, "sweep": [ { "cfg": { "lr": 0.006, "weight_decay": 0.0, "k": 1.0, "delta": 0.02 }, "mean": 0.000786285956564825, "std": 0.00016373570475228794 }, { "cfg": { "lr": 0.006, "weight_decay": 0.0, "k": 1.0, "delta": 0.05 }, "mean": 0.0007858229582780041, "std": 0.0001619668831625854 }, { "cfg": { "lr": 0.006, "weight_decay": 0.0, "k": 1.0, "delta": 0.12 }, "mean": 0.0007801908141118474, "std": 0.000154764005468824 } ], "full": { "cfg": { "lr": 0.006, "weight_decay": 0.0, "k": 1.0, "delta": 0.12 }, "mean": 0.0007801908141118474, "std": 0.000154764005468824, "per_seed": [ 0.0007548112771473825, 0.0008833256433717906, 0.0005147535703144968, 0.0009292386239394546, 0.0005692947888746858, 0.0008575883111916482, 0.0007604749407619238, 0.0009720393572933972 ], "n": 8, "signatures": [ { "mean_V": 0.014306977391242981, "mean_exact_u_norm": 0.026661492884159088, "mean_applied_u_norm": 0.0, "mean_hold_error": 0.026661492884159088, "event_rate_per_sequence_step": 0.0, "max_hold_error": 0.073698490858078 }, { "mean_V": 0.04791915416717529, "mean_exact_u_norm": 0.051865410059690475, "mean_applied_u_norm": 0.0, "mean_hold_error": 0.051865410059690475, "event_rate_per_sequence_step": 0.0, "max_hold_error": 0.10102833807468414 }, { "mean_V": 0.019514720886945724, "mean_exact_u_norm": 0.032075561583042145, "mean_applied_u_norm": 0.0, "mean_hold_error": 0.032075561583042145, "event_rate_per_sequence_step": 0.0, "max_hold_error": 0.08094639331102371 }, { "mean_V": 0.01886623166501522, "mean_exact_u_norm": 0.030544035136699677, "mean_applied_u_norm": 0.0, "mean_hold_error": 0.030544035136699677, "event_rate_per_sequence_step": 0.0, "max_hold_error": 0.08747458457946777 }, { "mean_V": 0.021111171692609787, "mean_exact_u_norm": 0.03352213650941849, "mean_applied_u_norm": 0.0, "mean_hold_error": 0.03352213650941849, "event_rate_per_sequence_step": 0.0, "max_hold_error": 0.07995948940515518 }, { "mean_V": 0.026953421533107758, "mean_exact_u_norm": 0.038697995245456696, "mean_applied_u_norm": 0.0, "mean_hold_error": 0.038697995245456696, "event_rate_per_sequence_step": 0.0, "max_hold_error": 0.07766351848840714 }, { "mean_V": 0.013531758449971676, "mean_exact_u_norm": 0.027224672958254814, "mean_applied_u_norm": 0.0, "mean_hold_error": 0.027224672958254814, "event_rate_per_sequence_step": 0.0, "max_hold_error": 0.07655420154333115 }, { "mean_V": 0.0649743378162384, "mean_exact_u_norm": 0.060073141008615494, "mean_applied_u_norm": 0.0, "mean_hold_error": 0.060073141008615494, "event_rate_per_sequence_step": 0.0, "max_hold_error": 0.09835705906152725 } ] } }, "comparison": { "delta_mean": 0.0, "idea_wins": 0, "n_pairs": 8, "per_seed_diffs": [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ], "p_value": 1.0, "mde": 0.0, "mde_rel_pct": 0.0, "verdict": "no measurable effect", "system_worked": false, "paired_diffs": [ 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0 ], "permutation_pvalue": 1.0 }, "mechanism_signature": { "prediction": "event rotation should preserve pair norms and bound hold error by delta", "observed": [ { "mean_V": 0.014306977391242981, "mean_exact_u_norm": 0.026661492884159088, "mean_applied_u_norm": 0.0, "mean_hold_error": 0.026661492884159088, "event_rate_per_sequence_step": 0.0, "max_hold_error": 0.073698490858078 }, { "mean_V": 0.04791915416717529, "mean_exact_u_norm": 0.051865410059690475, "mean_applied_u_norm": 0.0, "mean_hold_error": 0.051865410059690475, "event_rate_per_sequence_step": 0.0, "max_hold_error": 0.10102833807468414 }, { "mean_V": 0.019514720886945724, "mean_exact_u_norm": 0.032075561583042145, "mean_applied_u_norm": 0.0, "mean_hold_error": 0.032075561583042145, "event_rate_per_sequence_step": 0.0, "max_hold_error": 0.08094639331102371 }, { "mean_V": 0.01886623166501522, "mean_exact_u_norm": 0.030544035136699677, "mean_applied_u_norm": 0.0, "mean_hold_error": 0.030544035136699677, "event_rate_per_sequence_step": 0.0, "max_hold_error": 0.08747458457946777 }, { "mean_V": 0.021111171692609787, "mean_exact_u_norm": 0.03352213650941849, "mean_applied_u_norm": 0.0, "mean_hold_error": 0.03352213650941849, "event_rate_per_sequence_step": 0.0, "max_hold_error": 0.07995948940515518 }, { "mean_V": 0.026953421533107758, "mean_exact_u_norm": 0.038697995245456696, "mean_applied_u_norm": 0.0, "mean_hold_error": 0.038697995245456696, "event_rate_per_sequence_step": 0.0, "max_hold_error": 0.07766351848840714 }, { "mean_V": 0.013531758449971676, "mean_exact_u_norm": 0.027224672958254814, "mean_applied_u_norm": 0.0, "mean_hold_error": 0.027224672958254814, "event_rate_per_sequence_step": 0.0, "max_hold_error": 0.07655420154333115 }, { "mean_V": 0.0649743378162384, "mean_exact_u_norm": 0.060073141008615494, "mean_applied_u_norm": 0.0, "mean_hold_error": 0.060073141008615494, "event_rate_per_sequence_step": 0.0, "max_hold_error": 0.09835705906152725 } ], "mean_hold_error": 0.03758305567316711, "delta": 0.12, "confirmed": true }, "math_check": { "Vdot_observed": -0.006759319770566208, "Vdot_predicted": -0.0067593166955923385, "ratio": 1.0000004549237753, "passed": true }, "custom_track": null }